Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Forecasting of Wind Speed by Using Three Different Techniques of Prediction Models

Forecasting of Wind Speed by Using Three Different Techniques of Prediction Models Wind energy plays a major role in meeting the world’s growing power demand, due to which wind speed forecasting is essential for power system management, energy trading and maintaining the balance between consumption and generation for a stable electricity market. In this article, three different types of predicting techniques have been implemented for estimating wind speed by means of different meteorological parameters. Group method of data handling (GMDH), multi linear regression (MLR) and artificial neural network (ANN) models have been developed. For these models, data sets of 05 years (12 datasets from each year) were collected from the National Renewable Energy Laboratory (NREL). Five different types of input variables, which are ambient temperature (Ta), atmospheric pressure (Pr), wind direction (WD), relative humidity (RH) and precipitation (Pc) were selected as independent variables in all models. The collected wind speed (Wv) is selected as output or dependent variable. In this study, 48 sets of data were picked for training process and 12 datasets were selected for testing. The performances of models were examined using statistical parameters such as RMSE, MAPE and R2. MLR, GMDH and ANN techniques accurately performed with values of correlation coefficient (R) being obtained as 0.90552, 0.95542 and 0.97617 respectively. Comparative study of all models reveals that out of these three techniques, ANN performs the best. In the ANN model, the values of RMSE, MAE and R2 obtained were 0.17476, 0.12984 and 0.95210 respectively, which are optimal results when compared to those of other models. After ANN, GMDH performed better than MLR. Above analysis reveals that the wind speed was predicted with the highest accuracy by the neural technique. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Annals of Data Science Springer Journals

Forecasting of Wind Speed by Using Three Different Techniques of Prediction Models

Loading next page...
 
/lp/springer-journals/forecasting-of-wind-speed-by-using-three-different-techniques-of-NG9A4cPBaQ

References (48)

Publisher
Springer Journals
Copyright
Copyright © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2021
ISSN
2198-5804
eISSN
2198-5812
DOI
10.1007/s40745-021-00333-0
Publisher site
See Article on Publisher Site

Abstract

Wind energy plays a major role in meeting the world’s growing power demand, due to which wind speed forecasting is essential for power system management, energy trading and maintaining the balance between consumption and generation for a stable electricity market. In this article, three different types of predicting techniques have been implemented for estimating wind speed by means of different meteorological parameters. Group method of data handling (GMDH), multi linear regression (MLR) and artificial neural network (ANN) models have been developed. For these models, data sets of 05 years (12 datasets from each year) were collected from the National Renewable Energy Laboratory (NREL). Five different types of input variables, which are ambient temperature (Ta), atmospheric pressure (Pr), wind direction (WD), relative humidity (RH) and precipitation (Pc) were selected as independent variables in all models. The collected wind speed (Wv) is selected as output or dependent variable. In this study, 48 sets of data were picked for training process and 12 datasets were selected for testing. The performances of models were examined using statistical parameters such as RMSE, MAPE and R2. MLR, GMDH and ANN techniques accurately performed with values of correlation coefficient (R) being obtained as 0.90552, 0.95542 and 0.97617 respectively. Comparative study of all models reveals that out of these three techniques, ANN performs the best. In the ANN model, the values of RMSE, MAE and R2 obtained were 0.17476, 0.12984 and 0.95210 respectively, which are optimal results when compared to those of other models. After ANN, GMDH performed better than MLR. Above analysis reveals that the wind speed was predicted with the highest accuracy by the neural technique.

Journal

Annals of Data ScienceSpringer Journals

Published: Jun 1, 2023

Keywords: Group method of data handling; Multi linear regression; Artificial neural network; Wind speed; Wind energy; Prediction model

There are no references for this article.